Did a Comet Really Kill the Mammoths 12,900 Years Ago?

Did the planetary upheaval 12,900 years ago come from the heavens—or Earth?

The iconic Manicouagan meteor crater in Quebec is pictured in this handout photo taken March 14, 2013. The crater is one of the oldest known impact craters on Earth, still visible from space.


Why did mammoths, mastodons, and other mega-beasts vanish from North America?

Was it because:

1) humans killed them;

2) they couldn't hack the climate after the Ice Age ended; or

3) an exploding comet ignited continent-wide wildfires, sent hundred-mile-an-hour winds and tornadoes howling across the land, and shattered the North American ice sheet, while also maybe gouging out the Great Lakes?

Let's talk about option number three.

The idea that a comet struck Earth 12,900 years ago, at the beginning of a strange interlude of climate cooling called the Younger Dryas was first proposed in 2007. In the bitter scientific debate that has flared sporadically ever since, the latest evidence includes:

  • Tiny, glassy "spherules" of rock found in a Pennsylvania flowerbed by a woman who had seen a NOVA program about the comet hypothesis. In a paper that got wide coverage last week, Dartmouth researchers argue that those spherules were hurled to Pennsylvania by an impact in Quebec 12,900 years ago.

  • Traces of platinum deposited on the Greenland ice cap at about the same time. Harvard researchers argue that the platinum probably came from an extraterrestrial object—not a comet, however, but a rare type of iron-rich meteorite.

  • Spherules in Syria. In their latest paper, some of the original proponents of the impact hypothesis now say it deposited 10 million metric tons of spherules over an area of 20 million square miles, stretching from Syria through Europe to the west coast of North America.

Some opponents of the hypothesis—and there are many—want so badly for it to go away that they have attempted to declare it dead. "My only comment is that the pro-impact literature is, at this point, fringe science being promoted by a single journal," one of them, Nicholas Pinter of Southern Illinois University, said last week. The journal in question is Proceedings of the National Academy of Science.

Other researchers are trying to keep an open mind.

"Most people were trying to disprove this," said Wallace Broecker, a geochemist and climate scientist at Columbia University's Lamont-Doherty Earth Observatory. "Now they're going to have to realize there's some truth to it"—though maybe only a spherule or two.

Why Care About the Younger Dryas?

Even if you're inclined to let sleeping mammoths lie, this debate matters: It bears on the question of just how fragile Earth's climate is. Does it need an extraterrestrial whack to go haywire, or can it do that on its own?

The alleged mammoth-killing impact is also alleged to have triggered the Younger Dryas. At the time, 12,900 years ago, the continental ice sheets were in full retreat from the last Ice Age, and the planet was nearly as warm as it is now.

Suddenly, in a matter of decades, glacial temperatures returned, and the ice advanced again. The cold lasted 1,500 years, then ended even more suddenly than it had begun.

In the 1980s Broecker helped bring the Younger Dryas to wide attention. He explained the sudden cold snap with a mechanism that's internal to the climate system.

At the start of the Younger Dryas, he said, a conveyor belt of ocean currents that normally transports heat into the North Atlantic—the Gulf Stream is part of it—had gotten jammed by fresh meltwater flowing off the receding ice sheet. With no heat flowing north in the ocean, the North Atlantic region relapsed into bitter cold.

The Younger Dryas became the paradigm for the idea that Earth's climate was an intrinsically flighty creature, capable of shifting abruptly to a radically different state. That idea has made the prospect of future climate change even more worrisome.

Conceivably a comet might have triggered the Younger Dryas by helping to break up the ice sheet and send meltwater into the North Atlantic. It would have had to have been a big comet, with about a million times the energy of the bolide that excavated Meteor Crater in Arizona, which is three-quarters of a mile wide

Though proponents of a Younger Dryas impact say the comet may have exploded in air, leading to multiple impacts, no crater has yet been found.

Searching for Meteorite Metals

Mukul Sharma of Dartmouth and his colleagues now suggest at least one impact happened somewhere in Quebec. They base that conclusion on a very intricate analysis of some dirt from Yvonne Malinowski's backyard in Melrose, Pennsylvania.

A proponent of the impact hypothesis, a geological consultant named Allen West, sent Sharma sediment samples from half a dozen sites that he thought recorded evidence of a Younger Dryas impact, including Melrose. Sharma's graduate student, Yingzhe Wu, analyzed the sediment for osmium, a metal that is rare in Earth's crust but much more abundant in meteorites.

Like iridium, another member of the platinum group, osmium is considered a reliable signature of an extraterrestrial impact. The asteroid impact that wiped out the dinosaurs 65 million years ago, at the end of the Cretaceous Period, was ultimately traced to a large crater off the Yucatán Peninsula. But it was discovered first from an iridium-rich fallout layer that's clearly detectable all over the world.

Secrets of the Spherules

Sharma had no such luck with his Younger Dryas samples.

"In the sediment, there was no meteorite-derived osmium to be found, anywhere," he said. "That was remarkable." He and Wu didn't give up, however: Maybe the amount of osmium in the meteorite was just too small to detect in bulk sediment, they reasoned.

They decided to look next at individual spherules found in the Melrose dirt. Spherules are tiny glass beads, ranging in size from microscopic to a quarter of an inch or so. They form when rock or soil is somehow melted and then quickly cooled.

Meteorite impacts can form spherules, but so do volcanoes, lightning, and blast furnaces. Distinguishing impact spherules from the other kinds isn't easy.

The shape of the Melrose spherules—some were sort of like teardrops—showed that they cooled and solidified as they flew through the air, Sharma said. Distinctive minerals, including flecks of pure iron, showed that they'd formed at temperatures greater than 2,000 degrees Celsius (around 3,600 degrees Fahrenheit), comparable to the hottest part of a blast furnace.

There have never been any blast furnaces near Melrose. "So that convinced us they formed from a meteorite impact," Sharma said.

When the researchers analyzed individual spherules, however, they once again couldn't find osmium from a meteorite. But they did find, in the ratios of osmium and neodymium isotopes, a chemical signature resembling that of 1.5-billion-year-old rocks from Quebec.

That led them to conclude that the fireball from a meteorite impact in Quebec had dropped tiny bits of glass on Melrose.

Other scientists question that conclusion. Bill Glass of the University of Delaware, who has authored a treatise on impact spherules, said they tend to be almost perfectly spherical or teardrop shaped, whereas the Melrose ones are more irregular.

"I don't believe the Melrose 'spherules' are impact spherules," he wrote in an e-mail. "My guess is they are some kind of contamination."

Jay Melosh, a planetary scientist at Purdue University who also studies impacts—he was an early proponent of the Cretaceous dinosaur-killing impact theory—is bothered more generally by the Dartmouth researchers' reasoning.

"A lot of it is, 'Well, we don't understand how this is happening—must be an impact,'" Melosh said. "That's reasoning I see a lot. It's almost always wrong. Impacts are among the most rare features on the surface of the Earth. You need strong positive tests to prove them. The Younger Dryas doesn't pass."

A Stronger Test?

Even if the Melrose spherules were formed by an impact, there's no way to know whether it happened at the beginning of the Younger Dryas; the dirt layers at the site have not been precisely dated.

That's not a problem faced by researchers who study Greenland ice cores, which provide a precise record of climate change during the last ice age. In those cores, the annual snow layers can be counted like tree rings back to the Younger Dryas and beyond.

To put the Younger Dryas impact hypothesis to a rigorous test, Michail Petaev, Shichun Huang, Stein Jacobsen, and Alan Zindler of Harvard decided to look for iridium in one of the ice cores from Greenland. Their results appeared in the Proceedings of the National Academy of Sciences in July.  "We expected to find nothing," said Petaev.

And indeed they found next to no iridium.

But to the researchers' surprise, they found a pronounced spike in platinum that started exactly 12,900 years ago. Over the next 20 years or so the platinum concentration in the ice rose more than a hundredfold, then subsided again.

It's about the profile you'd expect, Petaev said, from dust settling out of the stratosphere after a meteorite impact, or perhaps a series of impacts.

Most meteorites contain about as much iridium as they do platinum. A rare kind called a magmatic iron meteorite, however, is platinum rich but iridium poor.

It's possible that a very small meteorite of that type happened to fall right on the part of Greenland that humans would extract an ice core from 12,900 years later. In other words, the platinum spike may be real, but it may be a coincidence that has nothing to do with mammoths or the Younger Dryas.

But it's also possible that there's a global platinum layer waiting to be discovered, like the iridium layer that proved there had been a Cretaceous impact.

To deposit as much platinum worldwide as it did on Greenland—around 30 parts per trillion—the meteorite would have had to be around half a mile across, the Harvard team calculates.

A rocky object that size would have left a substantial crater—but again, no Younger Dryas crater has been found. "I would bet on the event not being global," said Melosh.

The key test, though, will be whether the platinum spike turns up in Younger Dryas layers of Antarctic ice cores. Scientists are sure to be looking there soon.

Deus In or Ex Machina?

Opponents of the Younger Dryas impact hypothesis have raised all sorts of objections to it, besides the absence of a crater.

Perhaps the simplest is that there seems to be no need for such a deus ex machina to explain the drama that unfolded on Earth 13 millennia ago. Human hunting or climate change or both can explain the demise of mammoths and other Ice Age megafauna.

Internal lurches of Earth's climate machine suffice to explain the Younger Dryas itself—and besides, the ice-age climate record contains evidence of many other abrupt shifts similar to but earlier than the Younger Dryas. "You can't imagine every change had an extraterrestrial cause," Broecker said.

But the fact that an impact isn't needed at the Younger Dryas, or that some scientists may seem to want one too much, doesn't mean an impact didn't happen.

If an impact did happen at the Younger Dryas, it may just have amplified Earth's own internal sources of upheaval—extraterrestrial and terrestrial causes are not mutually exclusive. "The idea is that the system is drifting toward instability, but can't quite make it," Broecker said. "Then an impact comes along and it's like a knockout punch.

"But if it hadn't been for the impact, then the Younger Dryas would have just happened later,” he said. “It would have gone off by itself."

Researchers are only beginning, Broecker added, "to figure out what an impact did or didn't do. It's going to take a lot of people a lot of time."